'Click chemistry' synthesis of 1-(α-d-mannopyranosyl)-1,2,3-triazoles for inhibition of α-mannosidases

Article abstract of DOI:10.1016/j.carres.2015.01.004

Three new triazole conjugates derived from d-mannose were synthesized and assayed in in vitro assays to investigate their ability to inhibit α-mannosidase enzymes from the glycoside hydrolase (GH) families 38 and 47. The triazole conjugates were more sele

Full text of DOI:10.1016/j.carres.2015.01.004

Carbohydrate Research 406 (2015) 34e40
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
‘Click chemistry’ synthesis of 1-(
a- -mannopyranosyl)-1,2,3-triazoles
D
for inhibition of -mannosidases
a
a
,
*
b
b
c
Monika Pol aꢀ kov aꢀ , Rhiannon Stanton , Iain B.H. Wilson , Ivana Holkov aꢀ ,
Sergej Sest aꢀ k , Eva Machov aꢀ , Zuzana Jandov aꢀ , Juraj K oꢀ nꢁ a a
Institute of Chemistry, Center for Glycomics, Slovak Academy of Sciences, Dúbravska cesta 9, 845 38 Bratislava, Slovakia
Department für Chemie, Universit a€ t für Bodenkultur, A-1190 Wien, Austria
Department of Cellular and Molecular Biology of Drugs, Faculty of Pharmacy, Comenius University, Kalin ꢁc iakova 8, 832 32 Bratislava, Slovakia
ꢁ
a
a
a, c
a
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 27 October 2014
Received in revised form
Three new triazole conjugates derived from D-mannose were synthesized and assayed in in vitro assays
to investigate their ability to inhibit
3
a
potent towards GH38 mannosidases (IC50 values in the range of 0.5e6 mM towards jack bean
nosidase or Drosophila melanogaster lysosomal and Golgi
[
a
-mannosidase enzymes from the glycoside hydrolase (GH) families
8 and 47. The triazole conjugates were more selective for a GH47 -mannosidase (Aspergillus saitoi
1,2-mannosidase), showing inhibition at the micromolar level (IC50 values of 50e250 M), and less
-man-
-mannosidases). The highest selectivity ratio
IC50(GH38)/IC50(GH47)] of 100 was exhibited by the phenyltriazole conjugate. To understand structure-
a
9
January 2015
m
Accepted 11 January 2015
Available online 19 January 2015
a
a
Keywords:
Click reaction
activity properties of synthesized compounds, 3-D complexes of inhibitors with
built using molecular docking calculations.
a-mannosidases were
1-(
D-Mannopyranosyl)-1,2,3-triazole
a
-Mannosidase inhibitor
© 2015 Elsevier Ltd. All rights reserved.
Molecular docking
Glycoside hydrolase
1
. Introduction
simple glycoside mimetics has been supported by a high stability of
the glycosyl-triazole linkage under various conditions, including
14
In recent years the Cu(I)-catalysed azide-alkyne cycloaddition
reaction has became a powerful tool for an exclusive formation of
the 1,4-disubstituted 1,2,3-triazoles.
hydrolysis. They exhibited interesting biological activities as in-
1
5,16
17
hibitors of galectins,
glycogen phosphorylase, carbonic anhy-
1
e3
18,19
20
21
A broad applicability of the
drase
and O-GlcNAcase,
anticancer agents. Their cytotoxic activity and binding affinities
to lectins have also been investigated. Moreover, the ability of
or as antimycobacterial
and
4
22
23
reaction has been demonstrated in the field of material science,
chemical biology,5 medicinal chemistry,
6,7
and later also in the
24
field of carbohydrate chemistry. Many of the synthesized carbo-
hydrate mimetics were active against various biological targets.
One group of these mimetics derived from glycosyl azides has
been a subject of extensive investigation. The interest in these
these mimetics to inhibit the enzymatic activity of commercial
8
e13
25,26
glycosidases, such as sweet almond
b
-glucosidase,
yeast
a
b
-
-
2
6
25
glucosidase, Escherichia coli b-galactosidase, and bovine liver
2
5
galactosidase has been studied.
The processing of N-linked oligosaccharides by a-mannosidases
in the endoplasmic reticulum and Golgi complex is conserved in
2
7
plants and animals.
a-Mannosidases are enzymes that remove
Abbreviations: GMII, Golgi
nosidase II; dGMII, Drosophila melanogaster
domain of Drosophila melanogaster Co(II)-activated Golgi
combinant enzyme homolog of human Golgi -mannosidase II; LM, lysosomal
mannosidase; hLM, human lysosomal -mannosidase; bLM, Bos taurus lysosomal
mannosidase; dLManII, catalytic domain of Drosophila melanogaster lysosomal
mannosidase II, recombinant enzyme homolog of human lysosomal
nosidase; JBMan, jack bean -mannosidase (Canavalia ensiformis); AspMan,
1,2-mannosidase; hERManI, human endoplasmic reticulum
a
-mannosidase II; hGMII, human Golgi
-mannosidase II; dGMIIb, catalytic
-mannosidase IIb, re-
a-man-
2
8
a
non-reducing terminal mannose residues from glycoconjugates,
27
a
of which, there are two major homology-based classes. The
class I enzymes (the GH family 47) are present in the endoplasmic
reticulum and Golgi and are responsible for the initial trimming of
N-glycans to yield Man5ꢀ9GlcNAc structures; the class II enzymes
2
(the GH family 38) are more heterogeneous and encompass Golgi
enzymes that are involved in processing of hybrid and complex N-
a
a
a
a
-
-
-
a
a
-man-
a
Aspergillus saitoi
mannosidase I.
a
a-
glycans to result in GlcNAcMan
cytosolic enzymes involved in catabolism of glycoproteins. During
3 2
GlcNAc as well as lysosomal and
*
Corresponding author. Tel.: þ421 2 59410272.
E-mail address: Monika.Polakova@savba.sk (M. Pol ꢀa kov ꢀa ).
http://dx.doi.org/10.1016/j.carres.2015.01.004
008-6215/© 2015 Elsevier Ltd. All rights reserved.
0

M. Pol aꢀ kov aꢀ et al. / Carbohydrate Research 406 (2015) 34e40
35
cancer metastasis, the degree of branching of N-linked carbohy-
2. Results and discussion
drate structures has been correlated with malignancy and disease
progression through a disruption of normal intracellular in-
teractions and effectively concealing cancerous cells from immune
detection.29
An important feature of a compound playing the role of a po-
tential directed therapeutic application is its ability to inhibit only
2.1. Chemistry
The synthetic route to the target conjugates 6e8 is depicted in
Scheme 1. The -mannose azido building block 1 was prepared by
SnCl -catalysed reaction of peracetylated mannose with TMSN in
almost quantitative yield.
The application of the same procedure as reported in our pre-
D
4
3
2
3,38
some a-mannosidases, i.e. to exhibit as high as possible inhibition
selectivity towards the enzymes grouped to a single GH family or
even within a given GH family. A search for the selective inhibitors
is up to now a challenging area, because known potent inhibitors of
8
vious paper, i.e. coupling of 1 with alkynes 2ae2c at rt in DMF:H
2
O
(3:1) solvent mixture using copper (II) sulfate and sodium ascor-
bate, provided the 1,4-disubstituted 1,2,3-triazoles 3, 4 and 5 in
high yield. Subsequent removal of acetyl groups with K CO in
MeOH afforded the target conjugates 6, 7 and 8, respectively.
Their structures were identified by the presence of an olefinic
proton signal belonging to the 1,2,3-triazole moiety, which
appeared as a singlet at
simple reaction sequence provided glycoside mimetics suitable for
screening their selectivity and potency towards various
mannosidases.
3
9
human Golgi
a
-mannosidase II (hGMII) (EC 3.2.1.114, GH family 38)
are not sufficiently hGMII-selective to be interesting for the phar-
2
3
maceutical industry30 due to their unwanted co-inhibition of
39
catabolic lysosomal
a-mannosidase (LM) (EC 3.2.1.24, GH family
2
9
3
8).
Glycosyltriazole mimetics with
structure are suitable compounds for investigation of inhibitory
activity towards -mannosidases. We have previously reported the
synthesis of -mannosides differing in the aglycone, and examined
their inhibitory effect towards two -mannosidases, Drosophila
melanogaster homologs of human Golgi mannosidase II (dGMIIb)
1
a-linked mannose as the core
d
8.50e7.77 in the H NMR spectrum. This
a
a-
a
a
2.2. Biochemical evaluation and molecular modelling
31
and human lysosomal mannosidase (dLManII). Due to the pres-
ence of cis positioned hydroxyl groups, -mannose is an essential
part of the native substrate of GMII and its binding with the Zn
co-factor at the active site of GMII has been previously character-
D
In order to test the inhibitory activity of the synthesized
mannosides, a simple chromogenic assay using p-nitrophenyl
-mannoside was used with GH family 38 mannosidases. How-
2
þ
a
-
D
3
2,33
ized by crystallographic measurements.
We found that man-
ever, the GH family 47 enzyme (AspMan) does not accept aryl-
mannosides, therefore a normal phase HPLC-based assay with a
mannotetrasaccharide as the substrate had to be developed (see
Supplementary data). While sulfone derivatives 9e10 inhibited
nosides with a phenylalkylsulfonyl aglycone were weak GMII and
LM inhibitors (IC50¼1e3 mM). In addition, one of them, benzyl
a-D-
mannopyranosyl sulfone 9 was selective towards GMII. In our
ongoing research, we focused on a simple modification of the
aglycone and the phenylalkylsulfonyl function was replaced with
another hydrolytically stable triazolylphenylalkyl one. Three
mannose conjugates were synthesized and tested for their ability to
act as inhibitors of dGMIIb and dLManII. Moreover, the triazoles
only the
a-mannosidases from the GH family 38 (dGMIIb,
dLManII, JBMan) at the mM level, the triazole conjugates 6e8
showed opposite inhibitory activities (Table 1). They were more
potent and selective towards the
family 47 [IC50(AspMan)¼0.05e0.25 mM] and weaker inhibitors
of the -mannosidases from the GH family 38 [IC50(dGMIIb,
dLManII, JBMan)¼0.4e6 mM]. Within the GH family 38, all tri-
azole derivatives were more potent inhibitors of the plant
mannosidase [IC50(JBMan)¼0.4e1.1 mM] than of the animal Golgi
and lysosomal
-mannosidases [IC50(dGMIIb, dLManII)¼
.5e6 mM].
a-mannosidase from the GH
6
e8 and also the sulfones 9 and 10 (Scheme 1) were assayed to-
wards commercial enzymes, Canavalia ensiformis (jack bean)
mannosidase (JBMan) (EC 3.2.1.24, GH family 38) and Aspergillus
saitoi -1,2-mannosidase (AspMan) (EC 3.2.1.113, GH family 47), to
a
a
-
a
-
a
investigate their selectivity and inhibitory activity towards
mannoside-processing glycosidases. Based on high sequence sim-
ilarities between the active sites of human endoplasmic reticulum
a
0
3
4
a
-mannosidase I (hERManI) and AspMan, the latter was used as a
model enzyme for mammalian endoplasmic reticulum and Golgi
mannosidases from the GH family 47.
a
-
Table 1
Measured IC50 values for class I (AspMan) and class II (dGMIIb, dLManII and JBMan)
-mannosidases (in mM)
3
5e37
a
Compound
dGMIIb
dLManII
JBMan
AspMan
6
11.0ꢁ10ꢀ6
ꢀ4
ꢀ4i
ꢀ4j
>10a
OAc
O
OAc
O
OH
O
Swainsonine
5.0ꢁ10ꢀ
2.0ꢁ10
AcO
AcO
AcO
AcO
AcO
AcO
HO
HO
HO
1e5ꢁ10
a
b
ꢀ4
3.0ꢁ10ꢀ3
Mannostatin A
Kifunensine
1.5ꢁ10
2.4ꢁ10
c
0.12d
ꢀ4a
ꢀ5b
N3
N
N
N
N
5.2
2ꢁ10
N
2
e5ꢁ10
N
a
6
7
8
9
1
-mannosee
-D
n.i.f
n.i.f
n
n
1
5.0
5.0
6.0
2.0
1.5
5.0
0.5
5.0
0.4
1.1
0.4
2.0
1.9
0.05
0.1
3
n = 0
n = 1
n = 2
6 n = 0
7 n = 1
8 n = 2
4
5
2
a
0.25
g
g
h
n.i
n.i.
g
g
h
0
1.0
n.i.
a
b
c
d
e
f
36
OH
O
Ki value measured for hERMI by Vall eꢀ e et al.
2
2
b
c
HO
HO
HO
40
IC50 value measured for AspMan by Elbein et al.
41
K value measured for dGMII by Shah et al.
i
O
42
IC50 value measured by Kayakiri et al.
S
O
n
a mixture of a/bz95:5.
9
0
n = 1
n = 2
n.i.dno inhibition or inhibition <10% at 5 mM.
g
h
i
31
1
IC50 values measured by Pol ꢀa kov aꢀ et al.
n.i.dno inhibition or inhibition <10% at 1 mM.
43
Scheme 1. Reagents and conditions. (a) 2a, 2b or 2c, CuSO
4
, sodium ascorbate,
DMF:H O 3:1, 3 h, rt, 86e91%; (b) K CO , MeOH, 30 min, rt, 87e93%.
IC50 value measured by Kang et al.
j
44
2
2
3
IC50 value measured by Ogawa et al.

36
M. Pol aꢀ kov aꢀ et al. / Carbohydrate Research 406 (2015) 34e40
To better understand the selective binding and inhibitory ac-
tivities of the triazole conjugates towards the mannosidases from
both GH families 38 and 47, 3-D models of the inhibitor-enzyme
complexes were built by molecular docking techniques. The com-
4
5
pounds 6e8 were docked into dGMII (PDB ID: 3BLB), bLM (PDB
ID: 1O7D)46 and ERManI (PDB ID: 1FO3), which represent enzyme
models for the dGMIIb, dLManII and AspMan, respectively, used in
this paper. We also built 3-D complexes of sulfone derivatives 9e10
with dGMII, bLM and ERManI with the aim to compare the influ-
ence of the triazole and sulfone functional groups for binding of the
47
inhibitors to the a-mannosidases.
The triazoles 6e8, which showed the best binding properties
towards AspMan, have similar binding modes with ERManI
(
Fig. 1A). Their mannose moiety was bound at the bottom of the
2
þ
ERManI active site, interacting with Ca
ion, Glu663, Glu689,
Glu599, Phe659, Arg597 and a water molecule (Fig. 1B). The
binding poses of the mannose moiety of 6e8 were identical with
that found in a crystal structure of hERManI with a thio-
3
7
disaccharide substrate analogue. Due to the different lengths of
the arylalkyl triazole linkers of the inhibitors, apparent differ-
ences in the binding modes of these structural moieties were
found. The triazole ring of the best inhibitor 6 is oriented with its
polar N]N moiety to Arg597 and Leu525 to favour interactions
with the enzyme (Fig. 1B). These interactions were also found in
the crystal structure of nanomolar inhibitor kifunensine
3
6
(
IC50¼200 nM) with ERManI (PDB ID: 1FO3). The triazoles 7e8
with weaker inhibitory activities have a different binding mode
of the triazole ring (see a superposition of the docked structures
in Fig. 1A), which disfavours the interactions with Arg597 and
Leu525. The different accommodation of the triazole ring of these
inhibitors caused distinct positioning of their benzene moiety in
comparison with 6. The aromatic rings of all inhibitors were
placed in a pocket that consisted of polar (Glu396, Thr394,
Arg461) as well as hydrophobic amino acid residues (Ala460,
Phe329). Elongation of the aromatic linker (phenyl<-
phenylmethyl<phenylethyl) in 6e8 was accompanied with a
decrease of the inhibitory activity (50
mM>100 mM>250 mM,
Table 1). This is explained by less preferred interactions of the
aryltriazole moieties of 7 and 8 with Arg597, Leu525, Ala460 and
Arg461 as compared with our most potent inhibitor 6. The
importance of the triazole ring for inhibitory activities of 6e8 was
further supported by the additional biological assays of struc-
turally analogous sulfone derivatives 9e10. The compounds with
the sulfone moiety did not show any significant inhibition of
AspMan (IC50>1 mM).
As discussed above, the triazole and sulfone derivatives
exhibited opposite inhibitory activities towards the
a-man-
nosidases from the GH families 38 and 47. Taking into account our
previous results,31 it indicates that the replacement of the sulfone
function by the triazole ring did not significantly alter the potency
and selectivity of the inhibitors towards the a-mannosidases from
the GH family 38. By a superposition of the docked triazole deriv-
ative 7 with sulfone 9 both having benzyl function at the active site
of dGMII (Fig. 2), decreasing inhibitory activities of the triazole
derivatives 6e8 may be explained by less favoured interactions of
the triazole ring with Asp341, Tyr269 and Arg228. The distinct
positioning of the triazole ring also induced a different binding of
the aromatic moiety of the inhibitor compared with the sulfone
analogue.
Fig. 1. (A) A superposition of the compounds 6 (with carbons in green), 7 (in yellow)
3
6
and 8 (in magenta) docked in the active site of hERManI (PDB ID: 1FO3) ; (B) Selected
interactions (hydrogen bonds and hydrophobic contacts) between the inhibitor 6 (in
red) and hERManI. (For interpretation of the references to colour in this figure legend,
the reader is referred to the web version of this article.)
Additional docking of the triazole 7 into both dGMII and bLM
explained why desired selective binding of the triazoles towards
dGMII was not observed. The five best docking poses of 7 in dGMII
are compared with the docking poses in bLM (Fig. 3). The aryl tri-
azole moiety of 7 binds to hydrophobic regions (the sites B1 and B2
in Fig. 3), which are chemically similar in both enzymes. In contrast,
the site B3, which is different in both enzymes (a negatively
charged acidic dyad Asp270-Asp340 in dGMII versus polar neutral
dyad Asn262-Ser318 in bLM), and which could induce selective
binding of the inhibitors toward dGMII, was occupied by neither of
the triazoles.

M. Pol aꢀ kov aꢀ et al. / Carbohydrate Research 406 (2015) 34e40
37
Fig. 2. A superposition of the compound 7 with an analogous mannoside 9 with the
triazole group substituted by the sulfone group in the active site of dGMII (PDB ID:
4
5
3
BLB). For a comparison, interactions of triazole and sulfone group with active-site
amino acid residues of dGMII (Asp341, Tyr269 and Arg228) are visualized.
3
. Conclusion
Three triazole conjugates derived from
D-mannose showed
inhibitory activities against the a-mannosidase from the GH 47 at
the micromolar level, while their sulfone analogues were inactive.
The triazole conjugate 6 was the most potent inhibitor of the
enzyme from the GH family 47 [IC50(AspMan)¼50
mM]; its selec-
tivity is demonstrated by the IC50 ratio of 100 as compared to GH
family 38 members. The results from the molecular modelling
imply that the interactions of the polar N]N moiety of the inhibitor
triazole ring with Arg597 and Leu525 of the enzyme (hERManI) are
probably important for binding affinities of triazole conjugates.
The triazole conjugates were weak millimolar inhibitors of an-
imal and plant
mannose triazole structural moiety seems to be inappropriate
building block for inhibitors of the -mannosidases from the GH
a-mannosidases from the GH family 38. Thus, the D-
a
family 38. On the other hand, the sulfones only inhibited members
of this family, although with rather high IC50 values. Thus, further
refinements of their structures are required to explore their po-
tential as effective mannosidase inhibitors; furthermore, the
inhibitory activity of the triazoles in cell culture assays should be
tested to examine the potential of these compounds as inhibitors of
N-glycan processing in vivo.
Fig. 3. The best five docking poses for the triazole 7 found for dGMII and bLM. In both
cases the benzyl moiety of the inhibitor was found in two binding sites (B1 and B2,
which are similar in both enzymes). A binding site (B3), which is chemically different
4
. Experimental
(negatively charged Asp270-Asp340 dyad in dGMII vs polar neutral dyad Asn262-
4.1. General methods
Ser318 in bLM) and may be responsible for a selective binding of the inhibitor to-
wards dGMII, was empty. This may explain the lack of selectivity observed for the
triazole mannosides within the GH family 38 of a-mannosidases.
TLC was performed on aluminium sheets precoated with silica
gel 60 F254 (Merck). Flash column chromatography was carried out
on silica gel 60 (0.040e0.060 mm, Merck) with distilled solvents
Thermo Scientific Orbitrap Exactive instrument operating in posi-
(
hexane, ethylacetate, acetonitrile, methanol). Dichloromethane
tive mode. Inhibition assays for GH family 38 enzymes were per-
was dried (CaH ) and distilled before use. All reactions containing
2
5
1
formed using a microplate reader (Epoch, BioTek and Gen ™ data
sensitive reagents were carried out under an argon atmosphere. H
NMR and 13C NMR spectra were recorded at 25 C with a VNMRS
ꢂ
collection and analysis software), whereas a Shimadzu HPLC sys-
tem (equipped with an SIL-30AC injector, LC-30AD pump and a RF-
4
00 MHz Varian spectrometer, chemical shifts are referenced to
1
1
20AXS fluorescence detector) was used for the assays of the GH
either TMS (
to internal CDCl
d
0.00, CDCl
3
for H) or HOD (
OD (
3
d 4.87, CD OD for H), and
13
family 47 enzyme.
The conjugates 6e8 used in biological tests were lyophilized
before the use. p-Nitrophenyl a-D-mannopyranoside (pNP-Manp),
3
(d
77.23) or CD
3
d
49.15) for C. Optical ro-
ꢂ
tations were measured on a Jasco P2000 polarimeter at 20 C. High
resolution mass determination was performed by ESI-MS on a

38
M. Pol aꢀ kov aꢀ et al. / Carbohydrate Research 406 (2015) 34e40
jack bean
Sigma; Aspergillus saitoi
a
-mannosidase and alkynes 2aec were purchased from
35.4, 27.4 (Ph(CH
MS): m/z: calcd for [C24
2
)
2
), 20.9, 20.8 (2ꢁ), 20.7 (4ꢁCH
3
CO). HRMS (ESI-
]H : 504.19766. Found: 504.19788.
þ
a
1,2-mannosidase was purchased from
H
29
N
3
O
9
Prozyme and swainsonine and mannostatin A from Calbiochem. A
mixture of manno-tetrasaccharides (supplied by Dr. Machov aꢀ ;
4.2.2. Removal of acetyl protective groups. Synthesis of conjugates
(6e8)
Compound 3, 4 or 5 (0.2 mmol) was dissolved in methanol
2 3
(3 mL) and K CO (0.5 equiv) was added. The reaction mixture was
stirred at rt until disappearance of starting material (30 min), then
filtered and concentrated. The crude product was purified by col-
5
00
fluorescent tag and the major tetrasaccharide (Man
Man 1,2-Man-PA) was purified by reversed phase HPLC (Hyper-
clone 5
ODS C18, 250ꢁ4 mm; Phenomenex) followed by normal
m
g) was subject to pyridylamination in order to introduce a
a
1,2-Man 1,2-
a
a
m
phase HPLC (TSKgel Amide-80, 250ꢁ4.6 mm; Tosoh) analogous to
4
8
previously published procedures; the peaks containing the fluo-
rescent tetrasaccharide were verified by MALDI-TOF MS and up to
two mannose residues could be released from the substrate upon
umn chromatography (CH
3
CN:MeOH 15:1). Evaporation and
lyophilization gave the target compound 6, 7 or 8.
incubation with Aspergillus saitoi
most 1,2-linkage is resistant due to reduction of the reducing
terminus during pyridylamination).
a
1,2-mannosidase (the inner-
4.2.2.1. 1-
White powder (57.2 mg, 93%). [a]
a
-
D
-Mannopyranosyl-4-phenyl-1,2,3-triazole
þ82 (c 0.9, MeOH); lit
þ98.0 (c 1.34, MeOH). H NMR (400 MHz, CD OD): 8.50 (s,
H, CHN), 7.85e7.81 (m, 2H, C ), 7.46e7.33 (m, 3H, C ), 6.08 (d,
1H, J1,2 2.7 Hz, H-1), 4.76 (dd, 1H, J2,3 3.5 Hz, H-2), 4.12 (dd, 1H, J3,4
.5 Hz, H-3), 3.86 (dd, J5,6a 2.7 Hz, J6a,6b 12.5 Hz, H-6a), 3.81e3.76
(6).
3
9
a
D
1
[a
]
D
3
d
1
6
H
5
6 5
H
4
4
3
.2. Chemistry
8
13
.2.1. Synthesis of conjugates (3e5)
To a solution of azide 1 (0.1 g, 0.268 mmol) in DMF:H
:1) alkyne 2aec (1.1 equiv) was added followed by sodium
(m, 2H, H-4, H-6b), 3.40 (ddd, J5,6b 6.5 Hz, H-5); C NMR (100 MHz,
CD OD): 149.0 (NC]CH), 131.4, 130.0, 129.5, 126.8 (C ), 122.1
2
O (1.6 mL,
3
d
6 5
H
(NC]CH), 88.5 (C-1), 78.7 (C-5), 72.6 (C-3), 70.1 (C-2), 68.7 (C-4),
62.6 (C-6). HRMS (ESI-MS): m/z: calcd for [C14
308.12410. Found: 308.12421.
þ
ascorbate (0.042 g, 0.214 mmol) and Cu(II) sulfate (0.017 g,
.107 mmol). The reaction mixture was stirred at rt for about 3 h.
The reaction mixture was poured into satd. NH Cl (150 mL) and
H
17
N
3
O
5
]H :
0
4
extracted with EtOAc (3ꢁ25 mL). The organic extracts were com-
bined, washed with water, dried and concentrated. The crude
product was purified by column chromatography (hexane:EtOAc
4.2.2.2. 4-Benzyl-1-
White powder (57.2 mg, 89%). [
(400 MHz, CD OD): 7.82 (s,1H, CHN), 7.42e7.13 (m, 5H, C
(d, 1H, J1,2 2.7 Hz, H-1), 4.62 (dd, 1H, J2,3 3.5 Hz, H-2), 4.03e4.00 (m,
H, H-3, PhCH ), 3.76 (dd, J5,6a 2.7 Hz, J6a,6b 12.5 Hz, H-6a),
3.73e3.65 (m, 2H, H-4, H-6b), 3.25 (m, 1H, H-5); C NMR
(100 MHz, CD OD): 148.7 (NC]CH), 140.2, 129.6, 127.5 (C ),
123.7 (NC]CH), 88.2 (C-1), 78.5 (C-5), 72.5 (C-3), 70.0 (C-2), 68.6
a
-
D
-mannopyranosyl-1,2,3-triazole
(7).
1
a
]
D
þ63 (c 0.6, MeOH). H NMR
3
d
6 5
H
), 5.90
5
:2/1:1).
3
2
1
3
4
.2.1.1. 1-(2,3,4,6-Tetra-O-acetyl-
a
-
D
-mannopyranosyl)-4-phenyl-
þ58 (c 0.5, CHCl ).
). H NMR (400 MHz, CDCl ): 7.98 (s,
), 7.47e7.35 (m, 3H, C ), 6.08 (d,
1,2,3-triazole (3). Colourless oil (0.11 g, 90%). [
a
]
D
3
3
d
6 5
H
3
9
1
lit
[
a]
D
þ65.5 (c 1.01, CHCl
3
3
H
d
1
1
8
H, CHN), 7.87e7.83 (m, 2H, C
6
H
5
6
5
(C-4), 62.5 (C-6), 32.5 (PhCH
2
). HRMS (ESI-MS): m/z: calcd for
þ
H, J1,2 2.7 Hz, H-1), 6.02 (dd, 1H, J2,3 3.8 Hz, H-2), 5.98 (dd, 1H, J3,4
.8 Hz, H-3), 5.40 (t, J4,5 9.0 Hz, H-4), 4.40 (dd, J5,6b 5.4 Hz, J6a,6b
[C15
19
H N
3
O
5
]H : 322.13975. Found: 322.13983.
1
2.5 Hz, H-6b), 4.08 (dd, J5,6a 2.7 Hz, H-6a), 3.96 (ddd, H-5), 2.09,
4.2.2.3. 1-
White powder (58.3 mg, 87%). [
(400 MHz, CD OD): 7.77 (s,1H, CHN), 7.26e7.13 (m, 5H, C
(d, 1H, J1,2 2.5 Hz, H-1), 4.68 (dd, 1H, J2,3 3.5 Hz, H-2), 4.03 (dd, 1H,
3,4 8.7 Hz, H-3), 3.80e3.71 (m, H-4, H-6a, H-6b), 3.21 (ddd, 1H, J4,5
8.8 Hz, J5,6a 2.8 Hz, J5,6b 5.8 Hz, H-5), 3.01e2.97 (m, 4H, Ph(CH );
148.5 (NC]CH), 142.2, 129.5, 129.4,
), 123.3 (NC]CH), 88.3 (C-1), 78.2 (C-5), 72.5 (C-3), 70.1
a-D-Mannopyranosyl-4-(2-phenylethyl)-1,2,3-triazole (8).
13
1
2
CDCl
.07, 2.06, 2.04 (each s, each 3H, 4ꢁCH
):
170.6, 169.8, 169.7, 169.4 (4ꢁCH
29.8, 129.0, 128.7, 126.0, 119.9 (C , NC]CH), 83.7 (C-1), 72.3 (C-
3
CO); C NMR (100 MHz,
a
]
D
þ59 (c 0.6, MeOH). H NMR
3
d
3
CO), 148.4 (NC]CH),
3
d
6 5
H
), 5.94
1
5
6 5
H
), 68.9, 68.4 (C-2, C-3), 66.2 (C-4), 61.7 (C-6), 20.8, 20.7 (2ꢁ), 20.6
J
þ
(
4ꢁCH
3
CO). HRMS (ESI-MS): m/z: calcd for [C22
25
H N
3
O
9
]H :
2 2
)
1
3
476.16636. Found: 476.16676.
C NMR (100 MHz, CD
27.1 (C
3
OD): d
1
6 5
H
4
.2.1.2. 4-Benzyl-1-(2,3,4,6-tetra-O-acetyl-
a
-
D
-mannopyranosyl)-
(C-2), 68.4 (C-4), 62.5 (C-6), 36.5, 28.3 (Ph(CH
2
)
2
). HRMS (ESI-MS):
þ
1
,2,3-triazole (4). Colourless oil (0.11 g, 86%). [
a
]
D
þ30 (c 0.5, CHCl
3
).
),
m/z: calcd for [C16
H
21
N
3
O
5
]H : 336.15540. Found: 336.15558.
1
H NMR (400 MHz, CDCl
3
):
d
7.35e7.22 (m, 6H, CHN, C H
6
5
5
5
2
4
.96e5.91 (m, 3H, H-1, H-2, H-3), 5.35 (m, 1H, H-4), 4.33 (dd, J5,6b
.5 Hz, J6a,6b 12.5 Hz, H-6b), 4.12 (m, 2H, PhCH ), 4.04 (dd, J5,6a
.6 Hz, H-6a), 3.86 (ddd, H-5), 2.06, 2.04 (2ꢁ), 2.03 (each s, each 3H,
ꢁCH ): 170.5, 169.8 (2ꢁ), 169.4
CO); 13C NMR (100 MHz, CDCl
CO), 148.6 (NC]CH), 138.5, 128.8, 126.8, 121.9 (C , NC]
CH), 83.6 (C-1), 72.1 (C-5), 68.9, 68.4 (C-2, C-3), 66.2 (C-4), 61.7 (C-
), 32.22 (PhCH CO). HRMS (ESI-MS):
), 20.8 (2ꢁ), 20.7, 20.6 (4ꢁCH
m/z: calcd for [C23 ]H : 490.18201. Found: 490.18225.
4.3. Biochemistry
2
4.3.1. Enzyme preparation
The purification and characterization of recombinant Drosophila
melanogaster Golgi (dGMIIb) and lysosomal (dLManII) man-
3
3
d
(
4ꢁCH
3
6 5
H
2
7
nosidases was carried out as we described recently.
6
2
3
þ
a
-mannosidase assay31
27
H N
3
O
9
4.3.2. Class II
The supernatants of yeast expressing soluble forms of the
a
ꢂ
-
4
.2.1.3. 1-(2,3,4,6-Tetra-O-acetyl-
a
-
D
-mannopyranosyl)-4-(2-
þ28
3 3
c 0.5, CHCl ). H NMR (400 MHz, CDCl ): d 7.30e7.17 (m, 6H, CHN,
mannosidase were incubated with the substrate PNP-Manp at 37 C
for 2e3 h. The standard assay mixture consisted of 50 mM sodium
acetate buffer (pH 4.5 for JBMan, pH 5.2 for LManII or pH 5.8 for
GMIIb), 2 mM pNP-Manp (from 100 mM stock solution in DMSO),
phenylethyl)-1,2,3-triazole (5). Colourless oil (0.12 g, 91%). [
(
a]
D
1
Ph), 5.96 (d, 1H, J1,2 2.6 Hz, J2,3 3.7 Hz, H-2), 5.92e5.89 (m, 2H, H-1,
H-3), 5.36 (t, 1H, J3,4 9.1 Hz, H-4), 4.34 (dd, J5,6b 5.2 Hz, J6a,6b 12.5 Hz,
H-6b), 4.02 (dd, J5,6a 2.6 Hz, H-6a), 3.76 (ddd, H-5), 3.11e3.00 (m,
1e5
ml enzyme (supernatant of the culture medium) and in case of
GMIIb final 0.2 mM CoCl
sample contained no enzyme. The samples were prepared in trip-
licates. The reactions were terminated by the addition of 0.5 mL of
2
in a total reaction volume of 50 l. A blank
m
13
4
H, Ph(CH
2
)
2
), 2.08, 2.06, 2.05, 2.04 (each s, each 3H, 4ꢁCH
):
170.6, 169.8, 169.7, 169.4 (4ꢁCH
, NC]CH),
3.6 (C-1), 71.9 (C-5), 68.9, 68.5 (C-2, C-3), 66.2 (C-4), 61.6 (C-6),
3
CO);
C
NMR (100 MHz, CDCl
3
d
3
CO),
148.0 (NC]CH), 141.0, 128.6, 128.5, 126.3, 121.3 (C
6
H
5
2 3
100 mM Na CO and the absorbance was recorded at 410 nm
8
(spectrophotometer).

M. Pol aꢀ kov aꢀ et al. / Carbohydrate Research 406 (2015) 34e40
39
31
conformation of the mannose ring and in 3
4
.3.2.1. Inhibition assays . Inhibition assays of mannose conju-
S
1
one in the case of
gates 6e8 with dLManII and dGMIIb were carried out in the con-
ditions outlined above and the IC50 values were determined. Stock
concentrations of inhibitors were made up in DMSO to a concen-
hERManI. The receptor box for the docking conformational search
2
þ
was centred at the Zn ion co-factor (for (dGMII and bLM) and
2
þ
Ca ion (hERManI) bound at the bottom of the active site with a
ꢂ
tration of 100 mM and stored at ꢀ20 C. The inhibitory effect of
size of 39ꢁ39ꢁ39 Å using partial atomic charges for the receptor
2
þ
DMSO was tested for both enzymes. The 5% DMSO was selected as a
from the OPLS2005 force field except for the Zn and side chains
of His90, Asp92, Asp204, Arg228, Tyr269, Asp341 and His471
(analogous residues were selected for bLM). For these structural
fragments the charges were calculated at the quantum mechanics
level with the DFT (Density Functional Theory) method (M06-
maximal final concentration in the assay. This concentration causes
10% or 15% of inhibition of LM or GMIIb, eventually. For this reason
maximal concentration of the tested compounds in the reaction
was 5 mM and data were obtained in triplicate. The inhibitory effect
of the compounds was calculated by percentage of inhibition to-
ward a control sample containing the same concentration of DMSO.
Swainsonine and mannostatin A were used as standard mannose
inhibitors.
5
5,56
2X)
using hybrid quantum mechanics/molecular mechanics
(QM/MM) model (M06-2X/LACVP**:OPLS2005) with the QSite
5
7
program of the Schr o€ dinger package. For hERManI an analogous
2
þ
approach was used for the atomic charges calculations (Ca ion,
water molecules, side chains of Asn327, Glu570 and Glu663, se-
quences Glu397-Thr394, Gly459-Ser464, Phe329-Glu330, Ile333-
Arg334, Thr688-Glu689, Arg597-Glu599 and Leu525 were
treated in the QM part of the QM/MM calculations). The grid
maps were created with no Van der Waals radius and charge
scaling for the atoms of the receptor. Flexible docking in standard
(SP) precision was used. The partial charges of the ligands were
calculated at the DFT level (M06-2X/LACVP**). The potential for
nonpolar parts of the ligands was softened by scaling the Van der
Waals radii by a factor of 1.0 for atoms of the ligands with partial
atomic charges less than specified cut-off of 0.15. The 5000 poses
were kept per ligand for the initial docking stage with scoring
Inhibition of jack bean
a
-mannosidase activity was performed
l according to a pub-
The tested compounds 6e10 were prepared as
stock solutions in DMSO (100 mM) and data were obtained in
in microtiter plates in a final volume of 50
m
lished protocol.2
7,49
triplicate over a concentration range of 1e5 mM. 5
m
l of jack bean
ꢂ
a
-mannosidase was pre-incubated for 5 min at 20 C with the
inhibitor under evaluation. The reaction mixture contained 80 mM
sodium acetate buffer (pH 4.5) and 5 mM pNP-Manp. The absor-
bance of the reaction was measured at 405 nm using microplate
reader.
4.3.3. Class I
a-mannosidase assay
ꢀ1
As pNP-Manp is not a substrate for GH family 47 enzymes, an
HPLC-based assay was developed using a fluorescently-labelled
tetrasaccharide (for its preparation, see above). Approximately
window of 100 kcal mol for keeping initial poses; the best 400
poses were kept per ligand for energy minimization. The ligand
poses with rms deviations less than 0.5 Å and maximum atomic
displacement less than 1.3 Å were discarded as duplicates. The
post-docking minimization for 10 ligand poses with the best
docking score was performed and optimized structures were
1
m
g of manno-tetrasaccharide was incubated with 0.2
Aspergillus saitoi 1,2-mannosidase in the absence or presence of
.01, 0.1 or 1.0 mM inhibitor; all assays were performed at pH 5
40 mM ammonium acetate) in a volume of 10 l and incubated at
C overnight. Normal phase HPLC was then performed using a
ml (4 mU)
a
0
(
37
5
8
m
saved for subsequent analyses using the MAESTRO viewer of
ꢂ
the Schr o€ dinger package.
Shimadzu HPLC system (see above) to separate the smaller product
from the substrate using a Tosoh Amide 80 column; solvent A was
Acknowledgement
10 mM ammonium formate, pH 7, and solvent B was 95% (v/v)
acetonitrile/water. A gradient (1 mL/min) from 75% solvent B down
to 65% was applied over 10 min, before holding 65% solvent B for
This work was supported by the Slovak Research and Develop-
ment Agency (the project APVV-0484-12), Scientific Grant Agency
of the Ministry of Education of Slovak Republic and Slovak Academy
of Sciences (the project VEGA-2/0159/12), the Slovak State Pro-
gramme Project No. 2003SP200280203 and the Research &
Development Operational Programmes funded by the ERDF (Centre
of Excellence on Green Chemistry Methods and Processes,
CEGreenI, Contract No. 26240120001, and Amplification of the
Centre of Excellence on Green Chemistry Methods and Processes,
CEGreenII, Contract No. 26240120025) as well as by the Austria
Science Fund (FWF, grant P23607 to IBHW).
5
min and finally returning to the starting conditions. The conver-
sion of substrate to product was determined by integration of the
peak areas (fluorescence with excitation and emission at respec-
tively 320 and 400 nm).
4.4. Molecular modelling
The crystal structures of dGMII (PDB ID: 3BLB),45 bLM (PDB ID:
O7D)46 and hERManI (PDB ID: 1FO3) were used as 3-D enzyme
47
1
models of hGMII, hLM and AspMan for docking of synthesized
5
0,51
mannosides with the GLIDE program
of the Schr o€ dinger
Supplementary data
package.52 Protonation states of amino acid residues of enzymes
were calculated for the pH¼6.0±1 (dGMII), 4.5±1 (bLM) and
1H and 13C NMR spectra of the compounds 6e8 and examples of
HPLC data. Supplementary data related to this article can be found
at http://dx.doi.org/10.1016/j.carres.2015.01.004.
5
3
7
.0±1 (hERManI) by the Protein Preparation Wizard of the
Schr o€ dinger package. For docking with dGMII all molecules of
water at the active site of dGMII were deleted except one
(
WAT1820, numbering according to PDB ID: 3BLB). This water has
References
been shown to be conserved in crystal structures of dGMII either
with intact substrates or inhibitors.
hERManI all except of 8 molecules of water at the active site were
deleted (WAT705, 708, 709, 710, 711, 713, 803 and 925,
numbering according to PDB ID: 1FO3). In all docking calculations
the catalytic acid (Asp341 in dGM and Asp319 in the case of
3
2,33,45,54
For docking with
1. Tornoe CW, Meldal M. Chem Rev 2008;108:2952e3015.
2. Kolb HC, Sharpless KB. Drug Discov Today 2003;8:1128e37.
3. Moses JE, Moorhouse AD. Chem Soc Rev 2007;36:1249e62.
4. Lutz JF, Zarafshani Z. Adv Drug Deliv Rev 2008;60:958e70.
5. Baskin JM, Bertozzi CR. Qsar Comb Sci 2007;26:1211e9.
6. Tron GC, Pirali T, Billington RA, Canonico PL, Sorba G, Genazzani AA. Med Res
Rev 2008;28:278e308.
(
bLM) was modelled in the protonated form in accordance with
7
8
. Agalave SG, Maujan SR, Pore VS. Chem Asian J 2011;6:2696e718.
its catalytic role as a general acid. The mannoside ligands were
. Pol aꢀ kov ꢀa M, Bel ꢀa nꢁ ov aꢀ M, Miku ꢁs ov aꢀ K, Lattov aꢀ E, Perreault H. Bioconjug Chem
docked into the active site in the reactive skew boat B2,5
2011;22:289e98.

40
M. Pol aꢀ kov aꢀ et al. / Carbohydrate Research 406 (2015) 34e40
9. Marra A, Staderini S, Berthet N, Dumy P, Renaudet O, Dondoni A. Eur J Org Chem
34. Inoue T, Yoshida T, Ichishima E. Biochim Biophys Acta Protein Struct Mol Enzymol
2013:1144e9.
1995;1253:141e5.
10. Jogula S, Dasari B, Khatravath M, Chandrasekar G, Kitambi SS, Arya P. Eur J Org
35. Tempel W, Karaveg K, Liu ZJ, Rose J, Wang BC, Moremen KW. J Biol Chem
Chem 2013;2013:5036e40.
2004;279:29774e86.
1
1
1. Mukherjee C, Makinen K, Savolainen J, Leino R. Chem Eur J 2013;19:7961e74.
2. Wang GN, Andre S, Gabius HJ, Murphy PV. Org Biomol Chem 2012;10:
36. Vallee F, Karaveg K, Herscovics A, Moremen KW, Howell PL. J Biol Chem
2000;275:41287e98.
6
893e907.
37. Karaveg K, Siriwardena A, Tempel W, Liu ZJ, Glushka J, Wang BC, et al. J Biol
Chem 2005;280:16197e207.
1
1
1
3. Hsu TL, Hanson SR, Kishikawa K, Wang SK, Sawa M, Wong CH. Proc Natl Acad
Sci U S A 2007;104:2614e9.
4. Wilkinson BL, Bornaghi LF, Poulsen SA, Houston TA. Tetrahedron 2006;62:
38. Percec V, Leowanawat P, Sun HJ, Kulikov O, Nusbaum CD, Tran TM, et al. J Am
Chem Soc 2013;135:9055e77.
39. Schwardt O, Rabbani S, Hartmann M, Abgottspon D, Wittwer M, Kleeb S, et al.
Bioorg Med Chem 2011;19:64546473.
8
115e25.
5. Giguere D, Patnam R, Bellefleur MA, St-Pierre C, Sato S, Roy R. Chem Commun
006:2379e81.
2
40. Elbein AD, Tropea JE, Mitchell M, Kaushal GP. J Biol Chem 1990;265:
16. Tejler J, Skogman F, Leffler H, Nilsson U. J Carbohydr Res 2007;342:1869e75.
17. Bokor E, Docsa T, Gergely P, Somsak L. Bioorg Med Chem 2010;18:1171e80.
18. Salmon AJ, Williams ML, Maresca A, Supuran CT, Poulsen SA. Bioorg Med Chem
Lett 2011;21:6058e61.
15599e605.
41. Shah N, Kuntz DA, Rose DR. Biochemistry 2003;42:13812e6.
42. Kayakiri H, Takase S, Shibata T, Okamoto M, Terano H, Hashimoto M, et al. J Org
Chem 1989;54:4015e6.
1
2
2
2
2
2
9. Wilkinson BL, Innocenti A, Vullo D, Supuran CT, Poulsen SA. J Med Chem
43. Kang MS, Elbein AD. Plant Physiol 1983;71:551e4.
44. Ogawa S, Morikawa T. Eur J Org Chem 2000:1759e65.
45. Kuntz DA, Nakayama S, Shea K, Hori H, Uto Y, Nagasawa H, et al. Chembiochem
2010;11:673e80.
2
008;51:1945e53.
0. Li TH, Guo LN, Zhang Y, Wang JJ, Li ZH, Lin L, et al. Carbohydr Res 2011;346:
083e92.
1. Wilkinson BL, Long H, Sim E, Fairbanks AJ. Bioorg Med Chem Lett 2008;18:
265e7.
1
46. Heikinheimo P, Helland R, Leiros HKS, Leiros I, Karlsen S, Evjen G, et al. J Mol
6
Biol 2003;327:631e44.
47. Vallee F, Yip P, Lipari F, Sleno B, Romero P, Karaveg K, et al. Glycobiology
2000;10:1141.
48. Paschinger K, Hykollari A, Razzazi-Fazeli E, Greenwell P, Leitsch D, Walochnik J,
et al. Glycobiology 2012;22:300e13.
49. Kuntz DA, Ghavami A, Johnston BD, Pinto BM, Rose DR. Tetrahedron Asymmetry
2005;16:25e32.
2. Zhang JJ, Garrossian M, Gardner D, Garrossian A, Chang YT, Kim YK, et al. Bioorg
Med Chem Lett 2008;18:1359e63.
3. Hradilov aꢀ L, Pol aꢀ kov aꢀ M, Dvor aꢀ kov ꢀa B, Hajdúch M, Petru ꢁs L. Carbohydr Res
2012;361:1e6.
4. Azcune I, Balentova E, Sagartzazu-Aizpurua M, Santos JI, Miranda JI, Fratila RM,
et al. Eur J Org Chem 2013:2434e44.
2
2
5. Rossi LL, Basu A. Bioorg Med Chem Lett 2005;15:3596e9.
6. Dedola S, Hughes DL, Nepogodiev SA, Rejzek M, Field RA. Carbohydr Res
50. Friesner RA, Murphy RB, Repasky MP, Frye LL, Greenwood JR, Halgren TA, et al.
J Med Chem 2006;49:6177e96.
2
010;345:1123e34.
51. Friesner RA, Banks JL, Murphy RB, Halgren TA, Klicic JJ, Mainz DT, et al. J Med
Chem 2004;47:1739e49.
ꢁ
ꢀ
ꢁ
ꢀ
2
7. Nem ꢁc ovi ꢁc ov aꢀ I, Sestak S, Rendic D, Plskova M, Mucha J, Wilson IBH. Glycoconj J
2
013;30:899e909.
52. Glide, version 5.7. New York, NY: Schr o€ dinger, LLC; 2011.
53. Schr o€ dinger suite 2011 protein preparation wizard; Epik version 2.2. New York,
NY: Schr o€ dinger, LLC; 2011;
Impact version 5.7. New York, NY: Schr o€ dinger, LLC; 2011;
Prime version 3.0. New York, NY: Schr o€ dinger, LLC; 2011.
54. Kuntz DA, Zhong W, Guo J, Rose DR, Boons GJ. Chembiochem 2009;10:268e77.
55. Zhao Y, Truhlar DG. Theor Chem Acc 2008;120:215e41.
56. Zhao Y, Truhlar DG. Acc Chem Res 2008;41:157e67.
2
2
3
8. Winchester B. Biochem Soc Trans 1984;12:522e4.
9. Goss PE, Baker MA, Carver JP, Dennis JW. Clin Cancer Res 1995;1:935e44.
0. Bols M, Lopez O, Ortega-Caballero F. Glycosidase inhibitors: structure, activity,
synthesis, and medical relevance. In: Kamerling JP, editor. Comprehensive gly-
coscience. 1st ed., vol. 1. Oxford: Elsevier; 2007. p. 815.
1. Pol aꢀ kov ꢀa M, Sestak S, Lattova E, Petrus L, Mucha J, Tvaroska I, et al. Eur J Med
Chem 2011;46:944e52.
ꢁ
ꢀ
ꢀ
ꢁ
ꢁ
3
3
3
2. Shah N, Kuntz DA, Rose DR. Proc Natl Acad Sci U S A 2008;105:9570e5.
3. Zhong W, Kuntz DA, Ernber B, Singh H, Moremen KW, Rose DR, et al. J Am Chem
Soc 2008;130:8975e83.
57. Qsite, version 5.7. New York, NY: Schr o€ dinger, LLC; 2011.
58. Maestro, version 9.2. New York, NY: Schr o€ dinger, LLC; 2011.